1. Field of the Invention
The present invention relates to an image forming apparatus, an image processing apparatus, and a control method for them and, more particularly, to an image forming apparatus, an image processing apparatus, and a control method for them, which form an image by scanning a printhead having a plurality of printing elements on a print medium.
2. Description of the Related Art
A general example of an image output apparatus for a word processor, personal computer, facsimile apparatus, and the like is a printing apparatus which prints information such as desired characters and images on a sheet-like print medium such as a paper sheet or film. Such printing apparatuses use various printing methods. Among them, methods of forming an image on a print medium by making ink adhere to the print medium have been widely put into practice. As a typical example of such methods, the inkjet printing method has been known.
In such printing apparatuses, dots formed by printing elements sometimes vary in size and position, resulting in density unevenness in a printed image. In a serial type image forming apparatus, in particular, which performs printing by scanning a printhead in a direction different from the array direction of a plurality of printing elements, e.g., a direction perpendicular thereto, the above density unevenness sometimes appears as stripe unevenness on a printed image, resulting in a further deterioration in the quality of the printed image.
As a printing method for correcting such density unevenness, the multi-pass printing method is known. According to this technique, based on image data having undergone tone reduction processing (e.g., binarization), an image comprising one pixel or a line of pixels corresponding to one scan of printing elements is formed by dots formed by a plurality of different printing elements.
There has been proposed an image processing method in which when an image having undergone tone reduction processing like that described above is to be formed, the formation order and arrangement of the image are determined (see, for example, Japanese Patent Laid-Open No. 2000-103088). According to this technique, even if the registration of each scan varies upon tone reduction processing for each main scan, a deterioration in image quality due to density unevenness and the like can be suppressed.
More specifically, main scanning is performed for the same main scanning print area on a predetermined print medium a plurality of number of times by using different nozzle groups, and a binary image is formed for each main scan by an error diffusion method. When a binary image is generated by executing the error diffusion method for each main scan, the dots arranged within each main scan are high in dispersibility and uniform. Even if, therefore, the physical registration of the feeding amounts of a print medium or the positions of printing elements varies when an image is formed by a plurality of main scans, a change in graininess does not easily occur. In addition, since the correlation in dot arrangement between a plurality of main scans is low, even if registration variations occur, a change in dot coverage relative to the surface of a sheet is reduced, thereby considerably reducing density unevenness.
An error diffusion method is known as a means for converting multi-level input image data into a binary image corresponding to a dot print signal (or an image having the number of tone levels equal to or larger than two and smaller than the number of tone levels of input data). The error diffusion method implements pseudo tone expression by diffusing a binarization error which has occurred in a given pixel to a plurality of subsequent pixels.
In addition to this error diffusion method, a dither method is available as a means for converting multi-level input image data into binary image corresponding to a dot print signal (or an image having the number of tone levels equal to or larger than two and smaller than the number of tone levels of input data). The dither method implements pseudo tone expression by performing binarization by comparing a predetermined threshold matrix with multi-level input data. The dither method is known to be simpler than the error diffusion in terms of processing and hence be capable of high-speed processing.
A printhead used in the inkjet method or the like, which discharges liquid ink, has a very delicate structure, and hence sometimes suffers from a discharge failure as a dye or pigment, which is a dissolved substance in ink, sticks to ink orifices or the like of the printhead or a foreign substance such as dust adheres to ink orifices. This sometimes causes a printing failure in the printing apparatus.
Even in an image forming apparatus using printing elements based on a method (e.g., the electrophotographic method) other than the inkjet method, a printing failure sometimes occurs when printing elements fail or are damaged. When a printing failure occurs in printing elements in this manner, since they form no dots, a printed image does not satisfy a predetermined density. Furthermore, white stripes are formed along the main scanning direction.
There has been proposed a method of performing interpolation in an image forming apparatus based on the inkjet method when there are faulty nozzles. For example, there is available a technique in which when, for example, binarization is performed for a faulty nozzle position, an output value is forcibly set to 0, and an input value is diffused as an error to neighboring pixels by the error diffusion method (see, for example, Japanese Patent Laid-Open No. 2006-62088). According to this method, the density which should be printed by a faulty nozzle is interpolated by making neighboring nozzles of the faulty nozzle print more dots than those that should be printed in each main scan.
As another interpolation method, there is available a technique of assigning dots which should be printed by a faulty nozzle to other nozzles which form the same line by changing a mask table in multi-pass printing in which the positions of dots to be printed in each main scan are determined by the mask table (see, for example, Japanese Patent Laid-Open No. 2000-94662).
The following problems, however, arise in the above conventional faulty nozzle interpolation methods.
According to the method of interpolating the density, which should be printed by a faulty nozzle, by using neighboring nozzles of the faulty nozzle, it is possible to preserve the macroscopic density formed by each main scan. However, when attention is paid to a line in the main scanning direction which should be formed by a faulty line, since the number of dots to print the line cannot be interpolated, a white tripe is formed.
According to the method of assigning dots which should be formed by faulty nozzles to other nozzles which form the same line, the density of the line can be reproduced. This method, however, can be applied to only a case in which it is known in advance which nozzle is used to print which pixel of an image as in the case of multi-pass printing using a mask pattern. For this reason, the method cannot be applied to a case in which binarization is performed for each main scan so as to prevent density unevenness even with variations in registration. In addition, when a mask table is designed to optimize a dot pattern printed in each main scan, the presence of a faulty nozzle changes the mask table. As a consequence, the dot pattern in each main scan is not an optimized pattern.